Skip to content

L
linux
Switch branch/tag
  1. 01 Nov, 2006 8 commits
    • Hugh Dickins's avatar
      [POWERPC] Make high hugepage areas preempt safe · 96268889
      Hugh Dickins authored 
      Checking source for other get_paca()->field preemption dangers found that
open_high_hpage_areas does a structure copy into its paca while preemption
is enabled: unsafe however gcc accomplishes it.  Just remove that copy:
it's done safely afterwards by on_each_cpu, as in open_low_hpage_areas.
Signed-off-by: default avatarHugh Dickins <hugh@veritas.com>
Acked-by: default avatarDavid Gibson <dwg@au1.ibm.com>
Signed-off-by: default avatarPaul Mackerras <paulus@samba.org>
      96268889
    • Hugh Dickins's avatar
      [POWERPC] Make current preempt-safe · 5fe8e8b8
      Hugh Dickins authored 
      Repeated -j20 kernel builds on a G5 Quad running an SMP PREEMPT kernel
would often collapse within a day, some exec failing with "Bad address".
In each case examined, load_elf_binary was doing a kernel_read, but
generic_file_aio_read's access_ok saw current->thread.fs.seg as USER_DS
instead of KERNEL_DS.

objdump of filemap.o shows gcc 4.1.0 emitting "mr r5,r13 ... ld r9,416(r5)"
here for get_paca()->__current, instead of the expected and much more usual
"ld r9,416(r13)"; I've seen other gcc4s do the same, but perhaps not gcc3s.

So, if the task is preempted and rescheduled on a different cpu in between
the mr and the ld, r5 will be looking at a different paca_struct from the
one it's now on, pick up the wrong __current, and perhaps the wrong seg.
Presumably much worse could happen elsewhere, though that split is rare.

Other architectures appear to be safe (x86_64's read_pda is more limiting
than get_paca), but ppc64 needs to force "current" into one instruction.
Signed-off-by: default avatarHugh Dickins <hugh@veritas.com>
Signed-off-by: default avatarPaul Mackerras <paulus@samba.org>
      5fe8e8b8
    • Timur Tabi's avatar
      [POWERPC] qe_lib: qe_issue_cmd writes wrong value to CECDR · 302439d2
      Timur Tabi authored 
      Changed qe_issue_cmd() to write cmd_input to the CECDR unmodified.  It
was treating cmd_input as a virtual address and tried to convert it to
a physical address.
Signed-off-by: default avatarTimur Tabi <timur@freescale.com>
Signed-off-by: default avatarPaul Mackerras <paulus@samba.org>
      302439d2
    • Linas Vepstas's avatar
      [POWERPC] Use 4kB iommu pages even on 64kB-page systems · 5d2efba6
      Linas Vepstas authored 
      The 10Gigabit ethernet device drivers appear to be able to chew
up all 256MB of TCE mappings on pSeries systems, as evidenced by
numerous error messages:

 iommu_alloc failed, tbl c0000000010d5c48 vaddr c0000000d875eff0 npages 1

Some experimentation indicates that this is essentially because
one 1500 byte ethernet MTU gets mapped as a 64K DMA region when
the large 64K pages are enabled. Thus, it doesn't take much to
exhaust all of the available DMA mappings for a high-speed card.

This patch changes the iommu allocator to work with its own
unique, distinct page size. Although the patch is long, its
actually quite simple: it just #defines a distinct IOMMU_PAGE_SIZE
and then uses this in all the places that matter.

As a side effect, it also dramatically improves network performance
on platforms with H-calls on iommu translation inserts/removes (since
we no longer call it 16 times for a 1500 bytes packet when the iommu HW
is still 4k).

In the future, we might want to make the IOMMU_PAGE_SIZE a variable
in the iommu_table instance, thus allowing support for different HW
page sizes in the iommu itself.
Signed-off-by: default avatarLinas Vepstas <linas@austin.ibm.com>
Signed-off-by: default avatarBenjamin Herrenschmidt <benh@kernel.crashing.org>
Acked-by: default avatarOlof Johansson <olof@lixom.net>
Acked-by: default avatarStephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: default avatarPaul Mackerras <paulus@samba.org>
      5d2efba6
    • Andy Fleming's avatar
      [POWERPC] Fix oprofile support for e500 in arch/powerpc · dd6c89f6
      Andy Fleming authored 
      Fixed a compile error in building the 85xx support with oprofile, and in
the process cleaned up some issues with the fsl_booke performance monitor
code.

* Reorganized FSL Book-E performance monitoring code so that the 7450
  wouldn't be built if the e500 was, and cleaned it up so it was more
  self-contained.

* Added a cpu_setup function for FSL Book-E.  The original
  cpu_setup function prototype had no arguments, assuming that
  the reg_setup function would copy the required information into
  variables which represented the registers.  This was silly for
  e500, since it has 1 register per counter (rather than 3 for
  all counters), so the code has been restructured to have
  cpu_setup take the current counter config array as an argument,
  with op_powerpc_setup() invoking op_powerpc_cpu_setup() through
  on_each_cpu(), and op_powerpc_cpu_setup() invoking the
  model-specific cpu_setup function with an argument.  The
  argument is ignored on all other platforms at present.

* Fixed a confusing line where a trinary operator only had two
  arguments
Signed-off-by: default avatarAndrew Fleming <afleming@freescale.com>
Signed-off-by: default avatarPaul Mackerras <paulus@samba.org>
      dd6c89f6
    • Andy Fleming's avatar
      [POWERPC] Fix rmb() for e500-based machines it · e0da0dae
      Andy Fleming authored 
      The e500 core generates an illegal instruction exception when it tries
to execute the lwsync instruction, which we currently use for rmb().
This fixes it by using the LWSYNC macro, which turns into a plain sync
on 32-bit machines.
Signed-off-by: default avatarAndrew Fleming <afleming@freescale.com>
Signed-off-by: default avatarPaul Mackerras <paulus@samba.org>
      e0da0dae
    • Benjamin Herrenschmidt's avatar
      [POWERPC] Fix various offb issues · 441cbd8d
      Benjamin Herrenschmidt authored 
      This patch fixes a few issues in offb:

 - A test was inverted causing the palette hack to never work
(no device node was passed down to the init function)

 - Some cards seem to have their assigned-addresses property in a random
order, thus we need to try using of_get_pci_address() first, which will
fail if it's not a PCI device, and fallback to of_get_address() in that
case. of_get_pci_address() properly parsees assigned-addresses to test
the BAR number and thus will get it right whatever the order is.

 - Some cards (like GXT4500) provide a linebytes of 0xffffffff in the
device-tree which does no good. This patch handles that by using the
screen width when that happens. (Also fixes btext.c while at it).

 - Add detection of the GXT4500 in addition to the GXT2000 for the
palette hacks (we use the same hack, palette is linear in register space
at offset 0x6000).
Signed-off-by: default avatarBenjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: default avatarPaul Mackerras <paulus@samba.org>
      441cbd8d
    • Linus Torvalds's avatar
      Merge branch 'release' of master.kernel.org:/pub/scm/linux/kernel/git/aegl/linux-2.6 · d5b9b787
      Linus Torvalds authored 
      * 'release' of master.kernel.org:/pub/scm/linux/kernel/git/aegl/linux-2.6:
  [IA64] Correct definition of handle_IPI
  [IA64] move SAL_CACHE_FLUSH check later in boot
  [IA64] MCA recovery: Montecito support
  [IA64] cpu-hotplug: Fixing confliction between CPU hot-add and IPI
  [IA64] don't double >> PAGE_SHIFT pointer for /dev/kmem access
      d5b9b787
  3. 31 Oct, 2006 32 commits
GitLab Nexedi Edition | About GitLab | About Nexedi | 沪ICP备2021021310号-2 | 沪ICP备2021021310号-7